Shield tunneling machine tool bit defect detection system and method

Abstract

The invention discloses a shield tunneling machine tool bit defect detection system and method, and relates to the technical field of image defect detection.The shield tunneling machine tool bit defect detection system comprises an image acquisition unit arranged on a tool bit conveying path pair and used for acquiring tool bit surface images, and the image acquisition unit is divided into a surface acquisition module and a defect acquisition module; the surface acquisition module is arranged in multiple directions of the tool bit to acquire tool bit surface images, and the defect acquisition module acquires tool bit defect images at multiple angles based on tool bit surface defect positions; the image processing unit is used for identifying defect positions existing in the tool bit surface images and positioning the defect positions in the tool bit surface images; calculating a defect surface deviation angle; according to the method, the tool bit model is established and adjusted according to the tool bit defect image, so that the surface defect precision is calibrated, and the multi-angle tool bit defect image is acquired.

Description

Technical Field

[0001] The present invention relates to the technical field of image defect detection, and in particular to a shield machine cutter head defect detection system and method. Background Art

[0002] The shield machine cutter head is an important part of the shield machine, used for cutting and crushing rock and soil. The design and material selection of the cutter head directly affect the tunneling efficiency and construction quality of the shield machine. The cutter head usually adopts a modular design, especially the cutter part (such as carbide cutters), which will wear or damage after long-term operation. During the operation, the worn cutter is often removed and replaced according to the predetermined maintenance cycle or wear degree, and a new cutter module is replaced at the same time.

[0003] As an important part of the shield machine, the cutter head structure is kept on the machine during the entire tunnel excavation and even the project cycle. It will only be disassembled, repaired and replaced as a whole when there is significant wear or technical upgrade needs, and it will never be "buried" in the tunnel after the operation. Therefore, it is extremely important to systematically detect the cutter head of the shield machine and then respond to the cutter head according to the defect status of the cutter head.

[0004] Announcement No. CN116993731B discloses an image-based shield machine cutter head defect detection method, including: by performing Gaussian blur processing on the cutter head grayscale image under different Gaussian parameters, obtaining the first edge image corresponding to the change of edge pixels in the cutter head grayscale image, obtaining the change sensitivity of the second edge point in the cutter head grayscale image according to the number of edge pixels at the same position in several first edge images, and combining the distance between the second edge points in the local range, the change sensitivity and the difference between different parts to obtain an adaptive weight parameter, and performing image enhancement through the adaptive weight parameter to complete the shield machine cutter head detection. The enhancement effect of the cutter head grayscale image is improved, the defective area in the cutter head grayscale image is more prominent, the accurate detection of cutter head defects is improved, and the product quality of the shield machine cutter head is further improved.

[0005] The defect detection on the tool head surface is achieved by collecting the appearance of the tool head. However, the tool head surface image is affected by the image acquisition angle, which leads to errors in the detection of tool head surface defects and affects the detection accuracy of surface defects. Summary of the invention

[0006] One of the purposes of the present invention is to provide a shield machine cutter head defect detection system and method, in which a cutter head model is established through cutter head parameters, cutter head images are collected at different angles during the cutter head transportation process, the cutter head model is adjusted according to the state of the cutter head image, and the surface defect accuracy is corrected by multi-angle cutter head images.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a shield machine cutter head defect detection system, comprising: An image acquisition unit is arranged on the tool head conveying path to acquire the tool head surface image. The image acquisition unit is divided into a surface acquisition module and a defect acquisition module. The surface acquisition module is arranged on multiple directions of the tool head to acquire the tool head surface image. The defect acquisition module acquires the tool head defect images at multiple angles based on the defect position on the tool head surface. Image processing unit: Identify the defect position in the cutter head surface image, locate the defect position in the cutter head surface image, and calculate the defect surface deviation angle; Identify the defect status in the tool head defect image and calculate the defect range; A tool head model is established based on the tool head size, and the tool head model is adjusted according to the defect position in the tool head surface image. The defects in the tool head model are adjusted according to the defect data in the tool head defect image, and a tool head defect database is established; Based on the light position, tool head conveying speed and defect surface position, the optimal image acquisition position when the tool head is conveyed to different positions is calculated.

[0008] In one or more embodiments of the present invention, the number of surface acquisition modules is adjusted according to the tool head shape to be acquired, and the edge positions of the image acquisition ranges of the surface acquisition modules overlap, and the image acquisition range of multiple surface acquisition modules is combined to cover the surface of the tool head to be inspected.

[0009] In one or more embodiments of the present invention, a cutter head model for the batch of cutter heads is constructed using the cutter head design dimensions. When obtaining the defect positions in the cutter head surface image, the cutter head defects are simulated by adjusting the cutter head model. The cutter head model position is adjusted so that the cutter head model angle is consistent with the cutter head surface image acquisition angle, and defect positioning is performed using multiple cutter head surface images.

[0010] In one or more embodiments of the present invention, a three-dimensional coordinate system is established based on the tool head model, the tool head conveying posture placement surface is the three-dimensional coordinate bottom surface, the inclination angle of each surface of the tool head model in the three-dimensional coordinate system and the coordinates it contains are marked, the location of the defect in the tool head surface image is represented by the coordinates, and the defect surface in the tool head model where the defect is located is located according to the defect three-dimensional coordinates.

[0011] In one or more embodiments of the present invention, the offset angle of the cutter head during the conveying process is obtained through the cutter head surface image, the cutter head surface image is obtained by the surface acquisition module set at the position directly above the cutter head during the conveying process, the standard position of the cutter head during the conveying process is determined, the standard position reference point and the offset position reference point are obtained, and the positions of the two reference points in their respective images are determined using image processing tools or programming methods, and the offset angle is calculated using the following formula:i : ; in, Δx and Δy There are two reference points at x Axis and y The position difference on the axis.

[0012] In one or more embodiments of the present invention, the image acquisition environment of the defect acquisition module is obtained, including the light position, the tool head conveying speed and the defect surface deviation angle, and the light source irradiation environment simulation is performed: Establish a light source irradiation environment simulation model to determine the lighting angle, height, and interval distance; Determine the collection points of the defect collection module; Determine the trajectory of the cutter head.

[0013] In one or more embodiments of the present invention, the optimal image acquisition position is obtained based on the light source position, defect surface position and conveying dynamics: Step 1: Calculate the incident angle and irradiation intensity of the light on the defect surface through trigonometric functions. The angle between the light direction and the normal of the detection surface is α , when the light is perpendicular to the detection surface α=0° , the deviation angle between the actual normal direction of the defect surface and the normal direction of the detection surface is β , the positive or negative deviation angle depends on the deflection direction of the defect surface, and the incident angle is: , the irradiation intensity of the defect surface is: ,in, I 0 is the light intensity at vertical incidence, when i i =0° The light intensity reaches its maximum value when I 0 , i i When it increases, the irradiation light intensity decreases; Step 2: Determine the optimal image acquisition position based on the light source position, defect surface position and conveying dynamics, establish the coordinate system and parameter equation, and the light source position is L(x l ,y l ), the light direction is a constant, the light unit direction vector d l =(cosα,sinα), the light parameter equation is ,in represents the distance parameter along the ray; Mark the reference point in the projection line of the defect surface P d , the equation of the defect surface line is: ,in, For any point on the projection line of the defect surface, the normal vector is: , the reference point changes with time as: ;in, P d0 is the initial position of the reference point, v is the conveying velocity vector; Substituting the light parameter equation into the defect surface straight line equation is: ; Among them, the coordinates of the reference point at time t are ; The position where the light starts from the light source and intersects with the defect surface at time t is: .

[0014] In one or more embodiments of the present invention, the defects in the tool head model are further refined through the tool head defect image, and the size of the defect is calculated to obtain the area of ​​the defect surface. The defect size is determined based on the total area of ​​the defect surface, and the defect type is identified based on the defect state to determine the defect type and range.

[0015] The present invention also provides a tool head defect detection method, which is used in the above-mentioned automobile tool head defect detection system, and comprises the following steps: Collect the surface image of the cutter head, identify the surface image of the cutter head with defects, determine the defect location according to the surface image of the cutter head, and simulate the defect state through the cutter head model; Obtain the deviation angle of the defective surface of the tool head; Obtain the best image acquisition position based on the defect surface deviation angle; Capture the defect image of the tool head at the best image acquisition position; Improve the defect accuracy of the tool head model through tool head defect images; Calculate the defect range of the cutter head and enter the cutter head model into the cutter head defect database.

[0016] In one or more embodiments of the present invention, when performing optimal image acquisition, the lighting environment is simulated by means of the lighting position, the tool head conveying speed and the defect surface deviation angle.

[0017] Through the above technical solution, the present invention has the following beneficial effects: 1. The present invention establishes a tool head model and adjusts the tool head model according to the tool head defect image to calibrate the surface defect accuracy. By collecting tool head defect images from multiple angles, the influence of the environment on the surface defect accuracy during the tool head defect image collection process is reduced. The tool head model can adapt to all tool heads of the same type, and there is no need to model each tool head separately.

[0018] 2. By collecting tool head defect images from multiple angles, it is possible to obtain images of the tool head at different positions and with different light angles. The defect position images are displayed at multiple angles, and the state of the tool head model is adjusted according to the defect state, which can be used to calibrate the defect accuracy to reflect the defect accuracy information.

[0019] 3. After collecting the surface image of the tool head, the present invention analyzes the location of defects on the tool head surface, and collects multi-angle tool head defect images at the location where the surface defects exist, so as to collect at a certain location, shorten the collection time of the tool head surface defects while ensuring the accuracy of the collected defects, thereby improving the detection efficiency.

[0020] 4. During the image acquisition process, the best shooting position of surface defects is calculated according to the light position and its height, so as to ensure the best shooting effect during the detection process, to ensure the stability of the image acquisition picture, to reduce the influence of the light position on the defect position during the image acquisition process, and to improve the image accuracy after recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the detection system of the present invention; Figure 2 It is a schematic diagram of the standard position reference point and the offset position reference point of the present invention; Figure 3 The present invention is a flow chart of the detection method. DETAILED DESCRIPTION

[0022] The following will disclose multiple embodiments of the present invention with the accompanying drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. In other words, in some embodiments of the present invention, these practical details are not necessary. And if it is possible to implement, the features of different embodiments can be applied interchangeably.

[0023] Unless otherwise defined, all words (including technical and scientific terms) used herein have their usual meanings, which can be understood by those familiar with this field. Furthermore, the definitions of the above words in commonly used dictionaries should be interpreted in the content of this specification as consistent with the meanings of the relevant fields of the present invention. Unless otherwise clearly defined, these words will not be interpreted as idealized or overly formal meanings.

[0024] See also Figure 1-Figure 3 The present invention provides a shield machine cutter head defect detection system, which collects images of automobile cutter heads and identifies defects based on the cutter head images.

[0025] In one embodiment, the detection system comprises: An image acquisition unit is arranged on the tool head conveying path to acquire the tool head surface image. The image acquisition unit is divided into a surface acquisition module and a defect acquisition module. The surface acquisition module is arranged on multiple directions of the tool head to acquire the tool head surface image. The defect acquisition module acquires the tool head defect images at multiple angles based on the defect position on the tool head surface. Image processing unit: Identify the defect position in the cutter head surface image, locate the defect position in the cutter head surface image, and calculate the defect surface deviation angle; Identify the defect status in the tool head defect image and calculate the defect range; A tool head model is established based on the tool head size, and the tool head model is adjusted according to the defect position in the tool head surface image. The defects in the tool head model are adjusted according to the defect data in the tool head defect image, and a tool head defect database is established; Based on the light position, tool head conveying speed and defect surface position, the optimal image acquisition position when the tool head is conveyed to different positions is calculated.

[0026] In this embodiment, the defect status is obtained by processing the tool head surface image and the tool head defect image to locate the defect. Since the defects have different shapes, the accuracy of the defects is calculated, which can not only identify the defect position, but also obtain the defect accuracy, thereby improving the accuracy of the defect position.

[0027] A defect database is established for defective cutter heads, a virtual model can be established for the defective cutter heads, and the data in the database can be retrieved at any time.

[0028] During the image acquisition process, the deviation angle of the defective surface at the defective position of the tool head is used to obtain the optimal shooting position of the surface during the tool head transportation process to ensure the image quality of the defective surface of the tool head, thereby ensuring the accuracy of the tool head surface defect identification.

[0029] In one embodiment, the number of surface acquisition modules is adjusted according to the tool head shape to be acquired, and the edge positions of the image acquisition ranges of the surface acquisition modules overlap, and the image acquisition range of the combination of multiple surface acquisition modules covers the tool head surface to be inspected.

[0030] In this embodiment, multiple surface acquisition modules can locate the position of defects. When calculating the image acquisition position of the tool head defect, it can ensure that the defect is located at the optimal image acquisition position. The overlapping image acquisition of multiple surface acquisition modules can obtain the defect position at different angles, further accurately determining the location where the defect occurs.

[0031] In one embodiment, a tool head model for the batch of tool heads is constructed using tool head design dimensions. When obtaining the defect locations in the tool head surface image, the tool head defects are simulated by adjusting the tool head model. The tool head model position is adjusted so that the tool head model angle is consistent with the tool head surface image acquisition angle, and defect location is performed using multiple tool head surface images.

[0032] In this embodiment, the surface defects are positioned through multiple tool head images to accurately determine the defect location, thereby ensuring the accuracy of the calculation of the optimal image acquisition position for the defect, and further ensuring the acquisition quality of the tool head defect image. Adjusting the position of the tool head model so that it is consistent with the tool head surface image acquisition angle can further ensure the defect location accuracy.

[0033] In one embodiment, a three-dimensional coordinate system is established based on the tool head model, the tool head conveying posture placement surface is the three-dimensional coordinate bottom surface, the inclination angle of each surface of the tool head model in the three-dimensional coordinate system and the coordinates it contains are marked, the location of the defect in the tool head surface image is represented by the coordinates, and the defect surface in the tool head model where the defect is located is located according to the defect three-dimensional coordinates.

[0034] In this embodiment, a three-dimensional coordinate system of the model is established to represent the defect position. After the tool head model is subjected to defect simulation, the defect surface where the defect is located can be located. Since the inclination angle of each surface relative to the conveyor belt, i.e., the bottom surface, is constant, the inclination angle of the defect surface can be quickly acquired through the three-dimensional coordinate system of the defect.

[0035] Optionally, model three-dimensional coordinates and model positioning coordinates are established based on the cutter head model. The model positioning coordinates are the coordinates of the surface of the cutter head model. Since the bottom surface of the cutter head is not flat during the transportation process, when the defect position is obtained, the model positioning coordinates of the defect position can be used to obtain the model three-dimensional coordinates corresponding to the model positioning coordinates, so that the defect coordinates can be quickly identified.

[0036] In one embodiment, the offset angle of the cutter head during the conveying process is obtained through the cutter head surface image, the cutter head surface image is obtained by the surface acquisition module set at the position directly above the cutter head during the conveying process, the standard position of the cutter head during the conveying process is determined, the standard position reference point and the offset position reference point are obtained, and the positions of the two reference points in their respective images are determined using image processing tools or programming methods, and the offset angle is calculated using the following formula: i : ; in, Δx and Δy There are two reference points at x Axis and y The position difference on the axis.

[0037] In this embodiment, by calculating the offset angle of the cutter head during the conveying process, the defect surface of the cutter head model affected by the cutter head offset angle can be calculated during use, thereby ensuring the offset position accuracy of the defect surface and calculating the optimal image acquisition position of the defect surface.

[0038] Since not all the offsets of the various surfaces of the cutter head will be affected by the offset angle during the conveying process and when in use, it is necessary to determine whether the various surfaces of the cutter head are affected by the offset angle to choose whether to correct the offset angle. When the defective surface is not affected by the offset angle, the offset angle is not calculated.

[0039] In one embodiment, the image acquisition environment of the defect acquisition module is obtained, including the light position, the tool head conveying speed, and the defect surface deviation angle, and the light source illumination environment is simulated: Establish a light source irradiation environment simulation model to determine the lighting angle, height, and interval distance; Determine the collection points of the defect collection module; Determine the trajectory of the cutter head.

[0040] In this embodiment, a light source illumination environment simulation model is established, which can simulate the defect image acquisition environment of the tool head and analyze the optimal shooting position of the defect surface of the tool head at different positions under the same lighting environment, thereby ensuring the quality of the image of the defect surface collected by the defect acquisition module and further accurately detecting defects.

[0041] In one embodiment, the optimal image acquisition position is obtained based on the light source position, defect surface position and conveying dynamics: Step 1: Calculate the incident angle and irradiation intensity of the light on the defect surface through trigonometric functions. The angle between the light direction and the normal of the detection surface is α , when the light is perpendicular to the detection surface α=0° , the deviation angle between the actual normal direction of the defect surface and the normal direction of the detection surface is β , the positive or negative deviation angle depends on the deflection direction of the defect surface, and the incident angle is: , the irradiation intensity of the defect surface is: ,in, I 0 is the light intensity at vertical incidence, when i i =0° The light intensity reaches its maximum value when I 0 , i i When it increases, the irradiation light intensity decreases; Step 2: Determine the optimal image acquisition position based on the light source position, defect surface position and conveying dynamics, establish the coordinate system and parameter equation, and the light source position is L(x l,y l ), the light direction is a constant, the light unit direction vector d l =(cosα,sinα), the light parameter equation is ,in represents the distance parameter along the ray; Mark the reference point in the projection line of the defect surface P d , the equation of the defect surface line is: ,in, For any point on the projection line of the defect surface, the normal vector is: , the reference point changes with time as: ;in, P d0 is the initial position of the reference point, v is the conveying velocity vector; Substituting the light parameter equation into the defect surface straight line equation is: ; Among them, the coordinates of the reference point at time t are ; The position where the light starts from the light source and intersects with the defect surface at time t is: .

[0042] In this embodiment, since the defect surface moves with time, when the tool head passes through the detection area, 𝑃d(𝑡) and 𝑡 have a known linear relationship. For each moment 𝑡, the reference point 𝑃d(𝑡) of the current defect surface is calculated first; then the corresponding intersection point 𝑃 ​​is calculated using the above formula. int (𝑡); At the same time, since the basic angle of the defect surface is β, the incident angle It remains basically unchanged here (unless the local shape of the defect changes), so the theoretical irradiation intensity In practical applications, when the lighting system and image acquisition equipment are required to capture the "best image", it is necessary to first ensure that the defect surface is within the camera's field of view and that the intersection position coincides with the camera's best imaging surface.

[0043] Therefore, the determination of the optimal image acquisition position should satisfy: Lighting optimization: The incident angle 𝜃 at the intersection should be as small as possible to maximize the illumination intensity; Time-space matching: According to the conveying speed 𝑣, ensure that the camera moves to 𝑃 int The acquisition is triggered when (𝑡).

[0044] Calculate the intersection point between the defect surface and the light: Determine the geometric characteristics of the defect surface: including the size, shape and inclination angle of the defect surface.

[0045] Analyze the characteristics of the light source: including the type of light source (point light source, line light source or surface light source), light intensity distribution, and the position of the light source.

[0046] Apply ray tracing technology: Use ray tracing technology to simulate the process in which light is emitted from a light source, propagates through space, and finally intersects with the defect surface.

[0047] Calculate the intersection point of the light ray and the defect surface: Calculate the intersection point of the light ray and the defect surface based on the propagation path of the light ray and the position of the defect surface.

[0048] In another embodiment, the light and the defect surface are both regarded as three-dimensional objects, the light direction and the defect surface normal are represented by vectors, and the intersection point is solved using the line-plane intersection formula: ; The angle of incidence is still calculated by vector dot product: .

[0049] Combine the light position and direction with the real-time defect surface position of the tool head, and obtain a position 𝑃 that changes with time by solving the intersection of the light and the defect surface (straight line or plane) int (𝑡), at this position, the illumination reaches the maximum (or the incident angle is optimal), which can be regarded as the optimal image acquisition position.

[0050] In one embodiment, the defect in the tool head model is further refined through the tool head defect image, and the size of the defect is calculated to obtain the area of ​​the defect surface. The defect size is determined based on the total area of ​​the defect surface, and the defect type is identified based on the defect state to determine the defect type and range.

[0051] In this embodiment, by obtaining the defect type and range, the defect status can be judged, the defect type can be determined, the range of the defect can be calculated, and the defect can be simulated so that different defects can be counted and the causes of the defects can be analyzed.

[0052] The present invention also provides a tool head defect detection method, which is used in the above-mentioned automobile tool head defect detection system, and comprises the following steps: Collect the surface image of the cutter head, identify the surface image of the cutter head with defects, determine the defect location according to the surface image of the cutter head, and simulate the defect state through the cutter head model; Obtain the deviation angle of the defective surface of the tool head; Obtain the best image acquisition position based on the defect surface deviation angle; Capture the defect image of the tool head at the best image acquisition position; Improve the defect accuracy of the tool head model through tool head defect images; Calculate the defect range of the cutter head and enter the cutter head model into the cutter head defect database.

[0053] In one embodiment, when performing the optimal image acquisition position, the lighting environment is simulated by means of the lighting position, the tool head conveying speed and the defect surface deviation angle.

[0054] In summary, the technical solutions disclosed in the above embodiments of the present invention have at least the following advantages: 1. The present invention establishes a tool head model and adjusts the tool head model according to the tool head defect image to calibrate the surface defect accuracy. By collecting tool head defect images from multiple angles, the influence of the environment on the surface defect accuracy during the tool head defect image collection process is reduced. The tool head model can adapt to all tool heads of the same type, and there is no need to model each tool head separately.

[0055] 2. By collecting tool head defect images from multiple angles, it is possible to obtain images of the tool head at different positions and with different light angles. The defect position images are displayed at multiple angles, and the state of the tool head model is adjusted according to the defect state, which can be used to calibrate the defect accuracy to reflect the defect accuracy information.

[0056] 3. After collecting the surface image of the tool head, the present invention analyzes the location of defects on the tool head surface, and collects multi-angle tool head defect images at the location where the surface defects exist, so as to collect at a certain location, shorten the collection time of the tool head surface defects while ensuring the accuracy of the collected defects, thereby improving the detection efficiency.

[0057] 4. During the image acquisition process, the best shooting position of surface defects is calculated according to the light position and its height, so as to ensure the best shooting effect during the detection process, to ensure the stability of the image acquisition picture, to reduce the influence of the light position on the defect position during the image acquisition process, and to improve the image accuracy after recognition.

[0058] Although the present invention is disclosed in conjunction with the above embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the attached claims.

Claims

1. A shield machine cutter head defect detection system, characterized in that: include: An image acquisition unit is arranged on the tool head conveying path to acquire the tool head surface image. The image acquisition unit is divided into a surface acquisition module and a defect acquisition module. The surface acquisition module is arranged on multiple directions of the tool head to acquire the tool head surface image. The defect acquisition module acquires the tool head defect images at multiple angles based on the defect position on the tool head surface. Image processing unit: Identify the defect position in the cutter head surface image, locate the defect position in the cutter head surface image, and calculate the defect surface deviation angle; Identify the defect status in the tool head defect image and calculate the defect range; A tool head model is established based on the tool head size, and the tool head model is adjusted according to the defect position in the tool head surface image. The defects in the tool head model are adjusted according to the defect data in the tool head defect image, and a tool head defect database is established; Based on the light position, tool head conveying speed and defect surface position, the optimal image acquisition position when the tool head is conveyed to different positions is calculated.

2. A shield machine cutter head defect detection system according to claim 1, characterized in that: The number of surface acquisition modules is adjusted according to the shape of the tool head to be acquired, and the edge positions of the image acquisition ranges of the surface acquisition modules overlap, and the image acquisition ranges of the plurality of surface acquisition modules are combined to cover the surface of the tool head to be detected.

3. A shield machine cutter head defect detection system according to claim 2, characterized in that: The cutter head model of the batch of cutter heads is constructed according to the cutter head design dimensions. When the defect positions in the cutter head surface image are obtained, the cutter head defects are simulated by adjusting the cutter head model. The position of the cutter head model is adjusted to make the cutter head model angle consistent with the cutter head surface image acquisition angle, and defect positioning is performed using multiple cutter head surface images.

4. A shield machine cutter head defect detection system according to claim 3, characterized in that: The three-dimensional coordinates of the model are established based on the tool head model. The tool head conveying posture placement surface is the three-dimensional coordinate bottom surface. The inclination angle of each surface of the tool head model in the three-dimensional coordinates of the model and the coordinates it contains are marked. The position of the defect in the tool head surface image is represented by the coordinates, and the defect surface in the tool head model where the defect is located is located according to the defect three-dimensional coordinates.

5. A shield machine cutter head defect detection system according to claim 4, characterized in that: The offset angle of the cutter head during the conveying process is obtained through the cutter head surface image. The surface acquisition module set at the position directly above the cutter head during the conveying process obtains the cutter head surface image, determines the standard position of the cutter head during the conveying process, obtains the standard position reference point and the offset position reference point, uses image processing tools or programming methods to determine the positions of the two reference points in their respective images, and uses the following formula to calculate the offset angle θ : ; in, Δx and Δy There are two reference points at x Axis and y The position difference on the axis.

6. A shield machine cutter head defect detection system according to claim 5, characterized in that: Obtain the image acquisition environment of the defect acquisition module, including the light position, tool head conveying speed, and defect surface deviation angle, and simulate the light source irradiation environment: Establish a light source irradiation environment simulation model to determine the lighting angle, height, and interval distance; Determine the collection points of the defect collection module; Determine the trajectory of the cutter head.

7. A shield machine cutter head defect detection system according to claim 6, characterized in that: Get the best image acquisition position based on the light source position, defect surface position and conveying dynamics: Step 1: Calculate the incident angle and irradiation intensity of the light on the defect surface through trigonometric functions. The angle between the light direction and the normal of the detection surface is α , when the light is perpendicular to the detection surface α=0° , the deviation angle between the actual normal direction of the defect surface and the normal direction of the detection surface is β , the positive or negative deviation angle depends on the deflection direction of the defect surface, and the incident angle is: , the irradiation intensity of the defect surface is: ,in, I 0 is the light intensity at vertical incidence, when θ i =0° The light intensity reaches its maximum value when I 0 , θ i When it increases, the irradiation light intensity decreases; Step 2: Determine the optimal image acquisition position based on the light source position, defect surface position and conveying dynamics, establish the coordinate system and parameter equation, and the light source position is L(x l ,y l ), the light direction is a constant, the light unit direction vector d l =(cosα,sinα), the light parameter equation is in represents the distance parameter along the ray; Mark the reference point in the projection line of the defect surface P d , the equation of the defect surface line is: ,in, For any point on the projection line of the defect surface, the normal vector is: , the reference point changes with time as: ;in, P d0 is the initial position of the reference point, v is the conveying velocity vector; Substituting the light parameter equation into the defect surface straight line equation is: ; Among them, the coordinates of the reference point at time t are ; The position where the light starts from the light source and intersects with the defect surface at time t is: .

8. A shield machine cutter head defect detection system according to claim 7, characterized in that: The defect image of the tool head is used to further accurately identify the defects in the tool head model, calculate the size of the defect, obtain the area of ​​the defect surface, determine the defect size based on the total area of ​​the defect surface, and identify the defect type based on the defect state to determine the defect type and range.

9. A shield machine cutter head defect detection method, used in a shield machine cutter head defect detection system as claimed in any one of claims 1 to 8, characterized in that: The following steps are involved: Collect the surface image of the cutter head, identify the surface image of the cutter head with defects, determine the defect location according to the surface image of the cutter head, and simulate the defect state through the cutter head model; Obtain the deviation angle of the defective surface of the tool head; Obtain the best image acquisition position based on the defect surface deviation angle; Capture the defect image of the tool head at the best image acquisition position; Improve the defect accuracy of the tool head model through tool head defect images; Calculate the defect range of the cutter head and enter the cutter head model into the cutter head defect database.

10. A shield machine cutter head defect detection method according to claim 9, characterized in that: When the cutter head moves to the optimal image acquisition position, the lighting environment is simulated through the light position, cutter head conveying speed and its defect surface deviation angle.

Images